Developing cross-curricular rubrics:
Starting with the performance indicators identified for creative and practical problem solver, our next step is to build a analytical rubric that may be used in all classrooms asking students to demonstrate problem solving.
Analytical rubrics (example: Portfolio rubric Analytic) differ from holistic rubrics (example: Holistic General (Understanding)) by defining the criterion separately. Holistic rubrics describe the work by applying all the criteria at the same time to make an overall judgement.
There are reasons for using both types. Analytic rubrics provide students with clear feedback on what they are doing well and what needs improvement. However, they take more time to write and to use. Holistic rubrics provide little guidance to a student or parent about what a student may be struggling with in their work. However, they are easier to achieve inter-rater reliability (everyone who uses the guide, scores a piece of work the same way) and they are faster to use.
We are writing analytic rubrics for these Guiding Principles because we want students to be able to reflect upon their problem solving and see what skills they still lack. We are also writing general rubrics instead of task specific rubrics (Example here). General rubrics can be used across multiple products, content, or domains. These types of rubrics help students develop the transfer of specific skills into new contexts. General rubrics also lend themselves to sharing with students before an assessment is undertaken, as they do not contain a recipe for success, as can happen in task specific rubrics.
Criterion for a Good Rubric
(from Creating and Recognizing Quality Rubrics, by Judith Arter & Jan Chappuis)
The following (RubricforRubrics) is a guide for designing a good rubric. It contains several criteria:
Coverage and Organization: What counts in a student’s work?
- Covers the right content
- Does the content of the rubric represent the best thinking in the field about what it means to perform well on the skill or performance under consideration?
- Does the content of the rubric align with the content standards or learning targets intended to be assessed?
- Criteria are well organized
- Is the number of criteria appropriate for the complexity of the learning target?
- Are the descriptors for each criterion organized?
- Does the relative emphasis among criteria represent their relative importance?
- Is there minimal overlap between the criteria?
- Number of levels fit the learning target and intended use.
- Can users distinguish between levels?
Clarity: Does everyone understand what is meant?
- Levels defined well.
- Is each level of the rubric clearly defined?
- Do definitions rely on descriptive words and phrases rather than on non-specific words such as ‘outstanding’ or the use of counting or frequency measures?
- Could two independent teachers give the same rating?
- Is wording descriptive, not evaluative?
- Levels parallel
- Are features mentioned at one level, also mentioned at all other levels?
Common problems with rubrics
(from How to Create and Use Rubrics for Formative Assessment and Grading, by Susan Brookhart)
- Confounding the outcomes being measured (i.e. scoring more than one skill at a time without delineating them as separate skills (criterion))
- Scoring for extraneous features (e.g. neatness, color, length)
- Scoring by counting up parts or components rather than by looking for evidence of proficiency in the outcomes being measured.
- The rubric should not ensure that the requirements of the task are met, use a checklist or other tool for that. The rubric should inform the teacher and student about the qualities needed to demonstrate proficiency. Unless that standard refers to “providing 6 or more facts” don’t include these types of requirements in the rubric.
- Scoring for things students have not be cued to do or demonstrate.
- Scoring products rather than outcomes.
- The choice of criteria must be based upon on the learning goal not the task assigned to demonstrate understanding. If the criteria will not work for a different performance task, then the scoring may be overemphasizing the product at the expense of the learning target.
- This type of scoring doesn’t yield information that can support future learning, it provides information on what was done.
- Using the rubric to give an evaluative score (such as excellent, good, fair, or A, B, C)
- The rubric should provide descriptive scales not evaluative scales. They should allow the student and teacher to match performance to a description. It is a bridge between what can be seen in the student’s work and the judgement of that learning (meets, exceeds, etc). It is the explanation provided under each level that provides the feedback and the power of using a rubric for scoring.